Elderly patients have been presumed to be at increased risk of adverse renal effects of nonsteroidal anti-inflammatory drugs (NSAIDs). Documentation of renal effects of NSAIDs in this age group is poor. Moreover, elderly patients may have changed disposition of NSAIDs which could also influence effects on renal function. Though several studies have assessed the pharmacokinetics of some of the NSAIDs in elderly patients, more often than not they have not quantified disposition in terms of unbound drug. The importance of doing so is illustrated by studies with naproxen which demonstrated no change in disposition of total drug with age, but that clearance and volume of distribution of unbound naproxen in the elderly were about half that in young patients. Hence, most prior data of NSAIDs in the elderly are flawed. We propose two approaches to examining the Clinical Pharmacology of NSAIDs in the elderly: 1) a traditional pharmacokinetic/pharmacodynamic approach and 2) a pharmacoepidemiologic approach. In the former, we will use balance and clearance techniques to examine the effects of ibuprofen, sulindac and piroxicam on renal function. In so doing, we will also quantify the acute and steady state pharmacokinetics of total and unbound NSAID. With ibuprofen we will particularly examine the two enantiomers and with sulindac we will measure the parent drug, the active sulfide and the inactive sulfone. Using a pharmacoepidemiologic approach we will use the Regenstrief Data Base to address issues as to the prevalence of effects of NSAIDs on renal function, differences among NSAIDs, etc. This combined approach should yield on the one hand detailed information as to the disposition and effect of NSAIDs in the elderly and on the other hand quantitation of how these effects extrapolate to a large outpatient clinic.